The proposed cluster-randomized school-based study will examine the effectiveness of a program for the primary prevention of teen dating violence (TDV) and sexual violence (SV) perpetration among middle school male athletes. Engaging men and boys in preventing violence against women and girls is recognized by major global health organizations as a critical public health strategy. Coaching Boys into Men (CBIM) is a theory- and evidence-based program intended to alter gender norms that foster TDV/SV perpetration, promote bystander intervention (i.e., give boys skills to interrupt disrespectful and abusive behaviors they witness among peers), and reduce TDV/SV perpetration. Coaches receive a 60-minute training and biweekly check-in from the trainer to administer the intervention to their athletes via 12 mini-lessons conducted weekly throughout a sport season. In a randomized trial of program effectiveness among high school athletes (CDC CE001561- 01, PI Miller), male athletes receiving the program demonstrated increased positive bystander intervention behaviors and less abuse perpetration compared to controls. This proposal seeks to test the effectiveness of this program with younger male athletes in grades 6-8. The innovations are three-fold: (1) testing the efficacy of a novel TDV/SV prevention program for middle school male athletes; (2) training athletic coaches in TDV/SV prevention thus implementing primary prevention that does not rely on teachers or classroom time; and (3) changing gender norms integrated with the technique of a bystander intervention approach to reduce TDV/SV. This study is significant because we have too few evidence-based TDV/SV prevention programs for any age, there is only one other evidence-based TDV/SV prevention program for middle school, and we have no evidence-based TDV/SV programs for youth outside the classroom setting. The experimental design involves a 2-armed cluster-randomized-controlled trial in 26 middle schools in Pennsylvania. Schools will be randomly assigned to either intervention or control (standard coaching) condition. Coaches in intervention schools will receive CBIM training. Baseline surveys will be collected for all intervention and control site athletes entering grades 6-8 at the start of each sports seasons across Year 1 (Time 1; N=2200 athletes). Follow up surveys will be collected at the end of each sports season (Time 2). All participating athletes will be re-surveyed 12 months after baseline (Time 3; final N = 1980). Using generalized linear mixed models to account for clustering arising from randomizing at the level of schools, intervention effects on outcomes at Time 2 and at Time 3 will be assessed. The primary outcome is positive bystander behavior; secondary outcomes include increases in a) intentions to intervene when TDV/SV-related peer behaviors are witnessed, b) knowledge of what constitutes abusive behavior, and c) gender-equitable attitudes regarding relationship behaviors. Self-reported TDV/SV perpetration behaviors of athletes will also be assessed at both time points. Observations, interviews and focus group discussions will be conducted to document implementation processes.